Media Coverage of Hazard Events: Analyzing the Assumptions*

Abstract

Despite the widespread conviction that the mass media tend to “blow risks out of proportion,” the question has received little systematic attention. The mass‐media literature also presents the opposite argument, namely that the media minimize the kinds of reporting that might upset large‐scale capitalistic industries. Still other perspectives hold, on the one hand, that media reports tend to have more subtle anti‐technology effects, created by “even‐handed” coverage and on the other hand, that reporters give “establishment” figures the opportunity to “keynote” or “put the facts in perspective” in ways that exert a subtly pro‐technology effect. Drawing on a systematic sample of 128 hazard events developed by researchers at another university and factually summarized by Lexis‐Nexis, we analyze the effects of emotionalism on actual levels of coverage; a variety of analyses show that the only variables to exert significant effects are those that involve objective information, such as the number of casualties or the level of damage created. Given the argument that the effects of emotionalism can be conveyed in ways that are largely independent of the facts, e.g., by headlines, photographs, and “loaded” words, we next compare the emotionalism conveyed by factual summaries vs. the original stories and headlines. Even for incidents involving nuclear and/or toxic hazards, we find the net effect of the full stories is to lessen the emotionalism. Overall, the hypothesis receiving strongest support is that the “keynoting” helps create an overall impression that the “responsible authorities” are acting more responsibly than might be assumed on the basis of factual summaries alone. The widespread impression within the technical community, while understandable, may well have as much to do with the selective perceptions of scientists and engineers as with the actual pattern of media reporting.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 114 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.